116 ALASKAN GLACIER STUDIES therefore been applied not only to the valley feeder, but also to the stagnant, piedmont area that spreads westward until it partly unites with the Hubbard. In 1905 we assumed, with Russell and Gilbert, that the Variegated Glacier bulb was fed by two glaciers, the Variegated and Orange,1 but in 1906, on a journey to the supposed eastern tributary, now named the Orange Glacier, it was found that, while this glacier did actually coalesce with the Variegated for a short distance, the two were quite independent, and the stagnant piedmont area was entirely supplied and dominated by the Variegated Glacier. It has therefore been necessary to change the names applied in 1905 under the supposition that the piedmont area was supplied by two arms. The Variegated Glacier and its piedmont bulb presented many features of interest in 1905, especially the nature and distribution of its ablation moraines, but in 1906 it was absolutely transformed by a spasmodic advance. Therefore in both years it received considerable attention. The statement and discussion of the observations of 1905 and 1906, which are fully presented in the reports on the expedition of these years, are in the main so essential to a clear understanding of the changes observed in 1909, that it is necessary to restate the main points in the previously published description. Valley Portion in 1905.J Variegated Glacier descends in serpentine course from an unknown source among the mountains east of Mt. Seattle through a deep valley bordered by mountain walls greatly steepened by glacial erosion. We ascended this glacier to an elevation of 2100 feet, finding the ice everywhere smooth and easy to traverse (PL LUI), with only here and there a crevasse, or small group of crevasses, in the upper portion, but nowhere finding our progress impeded. In cross-section the glacier surface was flat, with a descent on either margin, increasing in steepness toward the valley mouth. The glacier grade varied considerably, but within the mountains was usually from 7 to 10 degrees, though near the valley mouth the grade became much less, and in some areas the ice surface was quite flat, and even locally had an upstream descent. Below the snow line, which lay just above the highest point which we reached (2100 feet), ablation was rapidly in progress increasing greatly in the lower portion, where both lateral and medial moraines appeared, while scattered patches of moraine and angular blocks were strewn over the surface. The evidence was complete that even well within its mountain valley the Variegated Glacier was only moderately active, while in the lower part of the valley portion the glacier was quite inactive. This inactivity of the lower portion was shown not only by the lack of crevassing, and the flat or undulating surface strewn with d6bris, but also by the presence of a marginal valley on either side with moraine-covered ice for one wall and the mountain side for the other. The smoothed and striated mountain walls, and the presence of moraine terraces on their slopes, on which only a few scattered annual plants had yet encroached, proved that at a very recent date this glacier had been more expanded and was then in contact with the mountain wall clear down to the end of the valley. The moraine on the glacier was mainly angular and evidently largely derived from rock falls from the mountain sides, which include a variety of rock types, including black i Tarr, R. S. and Martin, Lawrence, Glaciers and Glaciation of Yakutat Bay, Alaska, Bull. Amer. Geog. Sec.* Vol. XXXVlll, 1906, pp. 147-149. »See Tarr, R. S., The Yakutat Bay Region, Alaska, Professional Paper 64, U. S. Geol. Survey, 1909, pp. 46-58.